System and Method for Diagnosing Severity of Gastritis

1. A diagnostic system for determining a severity of gastritis in a subject, comprising: a processor programmed to: obtain a subject image of a gastric wall of the subject collected by an endoscope, the subject image showing a contrast between the gastric wall and a blood vessel of the subject; generate a subject abnormality image by comparing the subject image with a standard image representative of a healthy gastric wall, the standard image showing a contrast between the gastric wall and a blood vessel; compare the subject abnormality image with reference abnormality images representative of different severity levels of gastritis; and diagnose the subject as having a particular severity level of gastritis based on the comparison of the subject abnormality image with the reference abnormality images.

2. The diagnostic system according to claim 1 , wherein the processor is further programmed to determine abnormality scores for contrast areas of the subject image based on a measured luminance value.

3. The diagnostic system according to claim 2 , wherein the abnormality scores are further based on a position of each of the contrast areas.

4. The diagnostic system according to claim 3 , wherein: the processor is further programmed to calculate an overall abnormality score for the subject image based on the abnormality scores of the contrast areas; and the processor is programmed to diagnose the subject as having a particular severity level of gastritis based further on the overall abnormality score.

5. The diagnostic system according to claim 1 , wherein the standard image is a prior image of the gastric wall of the subject in a healthy state.

6. The diagnostic system according to claim 5 , wherein the processor is programmed to diagnose the subject as having a particular severity level of gastritis based further on a change in the blood vessel in the gastric wall of the subject over time.

7. The diagnostic system according to claim 1 , wherein the processor is further programmed to implement a deep learning model to generate the standard image from the subject image such that an abnormality in the subject image is omitted in the standard image.

8. The diagnostic system according to claim 1 , wherein the processor is further programmed to: generate the reference abnormality images by comparing reference images of different severity levels of gastritis with respective normal images representative of healthy gastric walls; and implement a deep learning model to generate the respective normal images representative of healthy gastric walls from the reference images such that an abnormality in the reference images is omitted in the normal images, respectively.

9. The diagnostic system according to claim 1 , wherein the reference abnormality images are selected based at least in part on a characteristic of the subject selected from the group consisting of age, race, and sex.

10. A method for determining a severity of gastritis in a subject, comprising: generating, via a processor, a subject abnormality image by comparing a subject image of a gastric wall of the subject collected by an endoscope with a standard image of a gastric wall, the subject image and the standard image each showing a contrast between the gastric wall and a blood vessel; comparing the subject abnormality image with reference abnormality images representative of different severity levels of gastritis; and diagnosing, via the processor, the subject as having a particular severity level of gastritis based at least in part on the comparison of the subject abnormality image with the reference abnormality images.

11. The method according to claim 10 , further comprising determining, via the processor, abnormality scores for contrast areas of the subject image based on a measured luminance value.

12. The method according to claim 11 , wherein the abnormality scores are further based on a position of each of the contrast areas.

13. The method according to claim 12 , further comprising calculating an overall abnormality score for the subject image based on the abnormality scores of the contrast areas, wherein the subject is diagnosed as having a particular severity level of gastritis based further on the overall abnormality score.

14. The method according to claim 10 , wherein the standard image is a prior image of the gastric wall of the subject in a healthy state.

15. The method according to claim 14 , wherein the diagnosing of the subject as having a particular severity level of gastritis is further based on a change in the blood vessel in the gastric wall of the subject over time.

16. The method according to claim 10 , further comprising generating the standard image from the subject image via a deep leaning model such that an abnormality in the subject image is omitted in the standard image.

17. The method according to claim 10 , further comprising: generating the reference abnormality images by comparing reference images of different severity levels of gastritis with respective normal images representative of healthy gastric walls; and generating the respective normal images representative of healthy gastric walls from the reference images via a deep learning model such that an abnormality in the reference images is omitted in the normal images, respectively.

18. The method according to claim 10 , wherein the reference abnormality images are selected based at least in part on a characteristic of the subject selected from the group consisting of age, race, and sex.

19. A computer-readable storage medium storing a computer-executable program that causes a computer to perform functions comprising: obtaining a subject image of a gastric wall of a subject collected by an endoscope, the subject image showing a contrast between the gastric wall and a blood vessel; generating a subject abnormality image by comparing the subject image with a standard image of a gastric wall, the standard image showing a contrast between the gastric wall and a blood vessel; comparing the subject abnormality image with reference abnormality images representative of different severity levels of gastritis; and diagnosing the subject as having a particular severity level of gastritis based at least in part on the comparison of the subject abnormality image with the reference abnormality images.

20. The computer-readable storage medium according to claim 19 , wherein the functions further comprise determining an abnormality score for the subject image based on measured luminance values and positions of contrast areas of the subject image.